Pterion body surface projection positioning method and device based on NCCT image data and storage medium

Abstract

The application discloses a method for locating the wing point on the body surface based on NCCT image data and belongs to the technical field of medical image processing. The method uses the NCCT sequence data of a routine image inspection at any time to reconstruct the shape of a skull, automatically and individually identifies the eyebrow bone features through corner point detection and Kmeans mean clustering, accurately determines the outer edge points of the eyebrow bone, adopts the method that the outer edge of the eyebrow bone is pushed forward to the direction of the extension line of the eyebrow bone by a curved surface distance of 3.5 cm and the midpoint of the malar arch is pushed upward by a linear distance of 4 cm, and the wing point is automatically and quickly located, which has important clinical significance.

Description

Technical Field

[0001] This invention belongs to the field of medical image processing technology, specifically relating to a pterion surface projection localization method based on NCCT image data. Background Technology

[0002] The pterion is an "H"-shaped suture formed by the intersection of the frontal, parietal, temporal, and greater wing of the sphenoid bones. It is located in the temporal fossa, two finger-widths above the midpoint of the zygomatic arch. The pterion is also known as the "sphenoid apex." It is approximately 3.5 cm posterior to the outer canthus of both eyes and 4 cm above the midpoint of the zygomatic arch. To visualize the pterion, place the thumb behind the frontal process of the zygomatic bone and the index and middle fingers of the other hand on top of the zygomatic arch, forming a triangle.

[0003] In clinical surgery, pterional craniotomy is an important surgical method, also known as the pterional approach. This involves removing portions of the frontal, temporal, and greater wing of the sphenoid bone to open the cranium and expose brain tissue from the lateral aspect. In surgical drainage of cerebral hemorrhage, the pterion serves as a golden reference point for locating the motor function areas (pre- and post-central gyri). Currently, the pterion is located by surgeons relying on experience and measuring it preoperatively with a ruler. Summary of the Invention

[0004] In view of the deficiencies of the existing technology, the present invention mainly provides a method for wing point surface projection localization based on NCCT image data.

[0005] To achieve the above-mentioned objectives, the present invention employs the following technical solution: a method for pterion surface projection localization based on NCCT image data, comprising the following steps:

[0006] The first step is to correct the head CT data;

[0007] The second step is to extract the overall head outline;

[0008] The third step is to extract the skull region based on the overall head contour obtained in step two to obtain the three-dimensional data of the entire head.

[0009] The fourth step is to locate the brow bone and determine the three-dimensional spatial coordinates (x, y) of the wing point;

[0010] Extract cross-sectional tomographic data of the brow bone region, and perform Harris-T corner detection on the outer edge of the bone structure of the section to determine the corners of all sections.

[0011] K-means clustering was performed on the obtained point clusters to obtain the outer edge point P1 of the brow bone.

[0012] Map the outer edge of the brow bone onto the body surface to obtain the body surface projection point P2 of the outer edge of the brow bone;

[0013] Based on the coordinates of the surface point P3, which is 3.5 cm below the sagittal plane where the surface projection point P2 is located, the x and y coordinates of the pterion are determined.

[0014] Fifth step, move the midpoint of the zygomatic arch upwards by 4cm to determine the z-axis coordinate of the wing point;

[0015] Step 6: Obtain the wingtip coordinates (x, y, z).

[0016] This invention, compared to existing technologies, has the following advantages and beneficial effects: The pterion surface projection localization method based on NCCT image data belongs to the field of medical image processing technology. This method utilizes NCCT sequence data from routine imaging examinations at any given time to reconstruct the skull shape. Through corner detection and K-means clustering, it automatically and individually identifies brow bone features, accurately determining the outer edge of the brow bone. It automatically and quickly locates the pterion by advancing a curved distance of 3.5cm from the outer edge of the brow bone towards the brow bone extension line and a straight distance of 4cm upwards from the midpoint of the zygomatic arch. This automated, intelligent, and individualized pterion localization replaces manual measurement, saving time and significantly increasing the accuracy of pterion localization. Attached Figure Description

[0017] Figure 1 This is a flowchart of a pterion surface projection localization method based on NCCT image data.

[0018] Figure 2 This is a standard CT scan image showing the effect of removing the bed slab.

[0019] Figure 3 This is a rendering of a portion of the fault after three-dimensional correction.

[0020] Figure 4 This is a partial tomographic image of the head contour.

[0021] Figure 5 This is a diagram showing the 3D reconstruction effect of the skull region and the definition of the 3D coordinate system.

[0022] Figure 6 It is a coordinate plane definition diagram.

[0023] Figure 7 This is a schematic diagram of the candidate layer for extracting the outer edge of the brow bone.

[0024] Figure 8 This is a schematic diagram of quadrant division of the candidate layer at the outer edge of the brow bone.

[0025] Figure 9 This is a diagram showing the detection results of the outer edge corner points of some candidate fracture bone structures.

[0026] Figure 10 This is a schematic diagram of the surface projection of the outer edge of the brow bone.

[0027] Figure 11 This is a schematic diagram illustrating the determination of the z-axis coordinates by the midpoint of the zygomatic arch. Detailed Implementation

[0028] The present invention will be further described in detail below with reference to specific embodiments and accompanying drawings. The following embodiments are merely illustrative of the present invention and should not be construed as limiting the invention.

[0029] The first step is three-dimensional image correction.

[0030] 1) Select brain CT scans with no lesions, no obvious abnormalities in bony features, and no angular deviation of the head during CT scans as the calibration benchmark, hereinafter referred to as standard data.

[0031] 2) Removing interference from standard data bed boards. The bed board removal operation employs conventional methods in the field, such as finding the largest connected component of the bony structure and removing the highlighted portion (e.g., ...). Figure 2 (As shown).

[0032] 3) Remove bedpan interference from the data to be corrected (brain CT scans of the pterion to be determined).

[0033] 4) The ITK 3D rigid registration tool was used to perform 3D calibration on the data to be calibrated (brain CT scans with pterion points to be determined). Figure 3 The correction diagrams for three of the faults are shown.

[0034] The second step is to extract the overall head contour.

[0035] The overall head contour is extracted from the brain CT scan image after correction in step one. The overall head contour refers to the scalp and all areas covered by the skin; this head contour ROI is tentatively termed vAirMask. The CT value (HU) of air is approximately -1000. Since the 3D correction algorithm itself normalizes pixels in areas near the head to a HU value of 0, areas with pixel values ​​greater than 0 are selected to determine the contour region of each plain CT image.

[0036] vAirMask i =1,vImage i >0

[0037] Where vImage is a plain CT image, and i represents the i-th slice. The value range of i is determined according to the number of slices in the specific CT image. For example, for a head CT image with a slice thickness of 5mm and a total of 32 slices, there are 32 plain CT images, i = 1, 2, 3...32.

[0038] Figure 4 The diagram shows the contour images of three sections. The solid vAirMask, i.e. the ROI region of the head contour, was obtained by layer-by-layer processing using three-dimensional morphology.

[0039] The third step is to extract the skull mask (grayscale value).

[0040] Within the vAirMask range obtained in step two, the skull portion is extracted. Based on the CT values ​​(HU) of bony structures ranging from 150 to 1000, the skull region is identified, and a three-dimensional spatial coordinate system is simultaneously determined. Figure 5 The image shows the three-dimensional reconstruction of the extracted skull.

[0041] The fourth step is to locate the brow bone and determine the spatial coordinates (x, y) of the wing point.

[0042] 1) Locate the section containing the brow bone in the xoz plane.

[0043] The sagittal plane of the brain is defined as the xoz plane, the transverse plane as the xoy plane, and the coronal plane as the yoz plane (e.g., Figure 6 As shown in the figure, using the longest slice (transverse section) of the brain tissue region in the sagittal plane as the benchmark, all slices within a 2.5cm range above it are taken. ROI As a candidate layer for extracting the outer edge of the brow bone (e.g. Figure 7 (As shown).

[0044] 2) Locate the quadrant where the brow bone is located.

[0045] The image data has undergone 3D correction. The brow bone area is located in the first and second quadrants of the cross-section. Because the left and right sides are symmetrical, in actual use, to locate the pterion on the desired side, the brow bone on the same side is detected. Taking the detection of the left pterion as an example, it is located in the second quadrant of the cross-section (e.g., Figure 8 (As shown).

[0046] Corner detection: The Harris-T corner detection algorithm is executed on the outer edge of the bone structure of the candidate fracture layers determined in step 1 (e.g., ...). Figure 9 As shown), find the corner points of all faults (e.g. Figure 9 (As shown).

[0047] 3) Perform K-means clustering on the point clusters (corner points) obtained in step 2 to obtain the coordinates P1(x1,y1,z1) of the final outer edge of the brow bone.

[0048] 5) Map the coordinates of the outer edge of the brow bone to the body surface. Determine the sagittal plane (xoz plane) where P1(x1,y1,z1) is located using its coordinates. Move P1(x1,y1,z1) along the x-direction of the z-axis in the sagittal plane and align it with the head contour. ROI The intersection of P1(x1,y1,z) and P2(x2,y2,z2) is projected onto the skin surface to obtain the surface projection of the outer edge of the brow bone.

[0049] 6) Move the x-axis line containing P2(x2,y2,z2)z downwards (z-axis) along the current layer vAirMask(:,:,iSlice) by 3.5cm to P3(x3,y3,z3). This determines the projected coordinates x3,y3 of the airfoil in the xoy plane (e.g., ...). Figure 10 (As shown).

[0050] Fifth step: Using the midpoint of the zygomatic arch, determine the z-axis coordinates of the pterion in three-dimensional space (e.g., Figure 11 (As shown).

[0051] Move the midpoint of the zygomatic arch upwards by 4cm to determine the z-axis coordinate of the pterion. The midpoint of the zygomatic arch is located at 0.4 times the length of the line connecting the nasal root and the occipital protuberance (reference point: nasal root). The methods for determining the nasal root and the occipital protuberance are based on existing techniques.

[0052] The sixth step is to combine the fourth and fifth steps to accurately determine the three-dimensional spatial coordinates of the wing points.

Claims

1. A method for pterion surface projection localization based on NCCT image data, characterized in that, Includes the following steps: The first step is to correct the head CT data; The second step is to extract the overall head contour based on the corrected brain CT scan obtained in the first step. The third step is to extract the skull region based on the overall head contour obtained in the second step to obtain the three-dimensional data of the entire head. The fourth step is to locate the brow bone and determine the coordinates (x, y) of the wing point in the three-dimensional (x, y, z) coordinate system. Extract cross-sectional tomographic data of the brow bone region, perform Harris_T corner detection on the outer edge of the bone structure of the section, and extract the corners of all sections; K-means clustering was performed on the obtained point clusters to obtain the outer edge point P1 of the brow bone. Map the outer edge of the brow bone onto the body surface to obtain the body surface projection point P2 of the outer edge of the brow bone; Based on the coordinates of the surface point P3, which is 3.5 cm below the sagittal plane where the surface projection point P2 is located, the x and y coordinates of the pterion are determined. Fifth step, move the midpoint of the zygomatic arch upwards by 4cm to determine the z-axis coordinate of the wing point; Step 6: Obtain the wingtip coordinates (x, y, z).

2. The method according to claim 1, characterized in that: In step one, the ITK 3D rigid registration tool is used to perform 3D correction on the brain CT data of the pterion to be determined using standard CT data.

3. The method according to claim 1, characterized in that: In step two, a solid head contour region is obtained by layer-by-layer processing using three-dimensional morphology.

4. The method according to claim 1, characterized in that: In step three, the skull portion is extracted within the vAirMask range obtained in step two to obtain three-dimensional head data.

5. The method according to claim 1, characterized in that... In step four, the specific steps for locating the brow bone and determining the three-dimensional spatial coordinates (x, y) of the wing point are as follows: (1) Locate the section where the brow bone is located in the plane. Using the longest section of brain tissue in the sagittal plane as a benchmark, all sections within a 2.5cm range above it were selected as candidate layers for extracting the outer edge of the brow bone. (2) Locate the quadrant where the brow bone is located. The candidate layer determined in (1) is divided into four quadrants, and the quadrant in which the brow bone is located is determined according to the position of the pterion to be determined. (3) Corner detection: The Harris-T corner detection algorithm is executed on the outer edge of the bone structure of the candidate fracture determined in step 1 to find the corners in all fracture target quadrant regions. (4) Perform K-means clustering on the corner point clusters obtained in (3) to obtain the coordinates P1(x1, y1, z1) of the outer edge of the brow bone; (5) Map the coordinates P1(x1, y1, z1) of the outer edge of the brow bone to the body surface to obtain the body surface projection P2(x2, y2, z2) of the outer edge of the brow bone; (6) Move point P2(x2, y2, z2) 3.5cm along the cross-sectional surface toward the cerebellum to P3(x3, y3, z). The projection coordinates of the pterion on the plane are x3, y3. Combined with the midpoint of the zygomatic arch moving 4cm toward the top of the head, the projection coordinates of the pterion on the Z plane are z3. In summary, the three-dimensional coordinates of the pterion in space are x3, y3, z3.

6. An electronic device comprising a memory and a processor, the memory storing a computer program executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the wing point surface projection localization method based on NCCT image data as described in any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps in the pterion surface projection localization method based on NCCT image data as described in any one of claims 1 to 5.

Images